Histopathology of Prostate Cancer

Abstract

This review focuses on histopathological aspects of carcinoma of the prostate. A tissue diagnosis of adenocarcinoma is often essential for establishing a diagnosis of prostate cancer, and the foundation for a tissue diagnosis is currently light microscopic examination of hematoxylin and eosin (H&E)-stained tissue sections. Markers detected by immunohistochemistry on tissue sections can support a diagnosis of adenocarcinoma that is primary in the prostate gland or metastatic. Histological variants of carcinoma of the prostate are important for diagnostic recognition of cancer or as clinicopathologic entities that have prognostic and/or therapeutic significance. Histological grading of adenocarcinoma of the prostate, including use of the 2014 International Society of Urological Pathology (ISUP) modified Gleason grades and the new grade groups, is one of the most powerful prognostic indicators for clinically localized prostate cancer, and is one of the most critical factors in determination of management of these patients.

The histopathological assessment of hematoxylin and eosin-stained tissue sections and, frequently, the detection of molecular markers by immunohistochemistry are critical for diagnosing, characterizing, and managing prostate cancer.

Establishment of a histopathological diagnosis of prostate cancer requires light microscopic examination of hematoxylin and eosin (H&E)-stained tissue sections. The most common prostatic parenchymal tissue samples examined in surgical pathology laboratories in the United States are, in order, 18-gauge needle cores, transurethral resections, radical prostatectomy specimens, open (simple or enucleation) prostatectomy specimens (uncommon), and fine-needle aspirates (rare). Needle core biopsies and fine needle aspirates may be used to diagnose metastatic prostate cancer.

Prostate cancer indicates a malignant neoplasm of the prostate. The vast majority of these malignant neoplasms are of epithelial origin and differentiation, and are carcinomas. These carcinomas are the subject of this review. There are rare malignant neoplasms in the prostate such as malignant mesenchymal neoplasms (sarcomas) and hematolymphoid neoplasms (lymphomas) of the prostate (Humphrey 2003a; Cheville et al. 2016; Ferry 2016); these nonepithelial neoplasms are not discussed here.

This review will focus on diagnostic histopathological and immunophenotypic attributes of prostatic carcinoma, including intraductal carcinoma, invasive adenocarcinoma, variants of prostatic carcinoma, and effects of treatment of prostatic carcinoma. Histopathological grading will also be presented.

INTRADUCTAL CARCINOMA

Intraductal carcinoma of the prostate is defined as an intra-acinar and/or intraductal neoplastic epithelial proliferation that fills large acini and prostatic ducts, with preservation of basal cells, and with formation of either solid or dense cribriform patterns, or a loose or micropapillary pattern with either marked nuclear atypia (with nuclear size 6× normal or larger) or comedonecrosis (Guo and Epstein 2006; Robinson et al. 2012; Epstein et al. 2016c). This entity is seen in ∼3% of needle core cases and 17% to 40% of radical prostatectomy cases; in these cases, it is typically admixed with invasive adenocarcinoma (McNeal and Yemoto 1996; Watts et al. 2013; Miyai et al. 2014). Only rarely is intraductal carcinoma detected in isolated form, without invasive adenocarcinoma, being diagnosed in 0.1% to 0.3% of needle core cases (Guo and Epstein 2006; Robinson et al. 2012; Watts et al. 2013).

Intraductal carcinoma is thought to represent a late event in prostate cancer evolution, and is typically associated with high-grade and high-stage invasive adenocarcinoma (Robinson and Epstein 2010). A current view is that most cases of intraductal carcinoma represent spread of invasive high-grade carcinoma into preexisting ducts and acini (Haffner et al. 2016). In a minority of cases, intraductal carcinoma could serve as a precursor proliferation (Miyai et al. 2014).

The main differential diagnostic considerations are high-grade prostatic intraepithelial neoplasia (PIN) and invasive but circumscribed high-grade prostatic acinar adenocarcinoma with cribriform or solid patterns (Wobker and Epstein 2016), as well as urothelial carcinoma with intraductal growth, and the intraductal component of ductal adenocarcinoma. Immunohistochemistry may be of use in addressing the differential diagnosis in selected cases. Loss of cytoplasmic PTEN expression favors intraductal carcinoma over high-grade PIN (Lotan et al. 2013; Morais et al. 2015) and complete absence of basal cells (as found with immunostains for p63 and/or high molecular weight cytokeratins) is indicative of invasive carcinoma rather than intraductal carcinoma. The prostate markers PSA, PSAP, prostein (P501S), and NKX3.1 will be positive in intraductal carcinoma, whereas GATA3, p63, and high-molecular-weight cytokeratin (detected by 34βE12 antibody binding) expression in neoplastic cells would be consistent with intraductal urothelial carcinoma. Ductal adenocarcinoma often has an intraductal component and these neoplastic cells are more often columnar in shape and pseudostratified with papillary architecture, compared with intraductal acinar carcinoma cells, but there may be overlap in their appearances.

Intraductal carcinoma is not assigned a Gleason grade (Epstein et al. 2016c). Intraductal carcinoma is linked to aggressive prostate cancer, as assessed by both pathologic and clinical outcome endpoints. It is usually associated with high-grade and high-volume carcinoma with a median Gleason score of 8 and T3 pathologic stage, with increased risk of extraprostatic extension, seminal vesicle invasion, and pelvic lymph node metastases (Robinson and Epstein 2010; Divatia and Ro 2016). The presence of intraductal carcinoma in needle biopsy tissue and transurethral resections is an independent indicator of early biochemical relapse and metastatic failure rate after radiation therapy (van der Kwast et al. 2012) and is associated with disease-specific survival (Kweldam et al. 2016). Intraductal carcinoma in radical prostatectomy tissue has been shown be an independent indicator of survival (Kimura et al. 2014) and to improve the capacity of nomograms to predict biochemical failure after surgery (O’Brien et al. 2011).

The diagnosis of isolated intraductal carcinoma in prostate needle biopsy tissue may prompt definitive treatment, although 10% of patients will have isolated intraductal carcinoma without invasive disease in the whole gland after radical prostatectomy (Robinson and Epstein 2010), such that rebiopsy may be also be an option.

INVASIVE ADENOCARCINOMA

Major and Minor Diagnostic Criteria

The foundation for the histologic diagnosis of adenocarcinoma of the prostate is the assessment for the three major criteria of glandular architecture, loss of basal cells, and nuclear features of the glandular lining cells (Table 1) (Totten et al. 1953; Mostofi et al. 1993; Humphrey et al. 2016a).

Table 1.

Criteria for diagnosis of prostatic adenocarcinoma

Major criteria

Architectural: Infiltrative small glands or cribriform glands too large or irregular to represent high-grade prostatic intraepithelial neoplasia (PIN)

Single-cell layer (absence of basal cells)

Nuclear atypia: nuclear and nucleolar enlargement

Minor criteria

Intraluminal wispy blue mucin (blue-tinged or basophilic mucinous secretions)

Pink amorphous secretions

Mitotic figures

Intraluminal crystalloids

Adjacent high-grade PIN

Amphophilic cytoplasm

Prostatic adenocarcinoma displays an abnormal architectural glandular pattern with disturbance of benign epithelial–stromal relationships. These alterations are best appreciated at low-power scanning magnifications. Many of the common growth patterns of prostatic adenocarcinoma are well illustrated by the International Society of Urological Pathology (ISUP) 2015 modified Gleason grading schematic diagram (Fig. 1), which presents five patterns of growth (Epstein et al. 2016b). The lowest grades, comprised of Gleason patterns 1 through 3, form grade group 1 in the new grade group scheme (Table 2). These malignant glands are single, separate, and well formed (Fig. 2). They may be crowded and nodular, as seen in Gleason patterns 1 and 2, which are characteristically detected in the transition zone. Gleason pattern 3 glands may be crowded or the glands may be haphazardly arranged and infiltrative into stroma (Fig. 2). The infiltration may be recognized as growth in between (Fig. 2) and around benign glands. Stromal infiltration is also seen in high-grade Gleason patterns 4 and 5 (and grade groups 2 to 5). Effacement of normal prostatic microanatomy with destruction of benign prostatic tissue and replacement by malignant glands can be observed with all grades. High-grade Gleason pattern 4 adenocarcinoma glands are cribriform (Fig. 3), poorly formed, fused, or glomeruloid (Fig. 1). High-grade Gleason pattern 5 adenocarcinoma consists of sheets of tumor (Fig. 4), individual cells, cords, linear arrays, and solid nests (Fig.1) (Gottipati et al. 2012; Epstein et al. 2016b; Humphrey et al. 2016a). A host response to stromal infiltration by adenocarcinoma, including a fibrogenic response and inflammatory response, is not usually evident on H&E-stained tissue sections. Uncommonly, there can be a sclerotic response to the invading glands. An inflammatory response is unusual, being seen in ∼10% of cases, and is usually lymphocytic. Also uncommon are acutely inflamed carcinoma glands, which usually show intraluminal neutrophils.

Figure 1.

Prostate cancer grades. 2015 modified Gleason grading schematic diagram according to the International Society of Urological Pathology (ISUP). Grade patterns 1 (top) through 5 (bottom). (From Humphrey et al. 2016a; reprinted, with permission.)

Table 2.

Grade groups

Grade group 1: Gleason score ≤6

Only individual discrete well-formed glands

Grade group 2: Gleason score 3 + 4 = 7

Predominantly well-formed glands with lesser component of poorly formed/fused/cribriform glands

Grade group 3: Gleason score 4 + 3 = 7

Predominantly poorly formed/fused/cribriform glands with lesser (>5%) component of well-formed glands

Grade group 4: Gleason score 4 + 4 = 8; 3 + 5 = 8; or 5 + 3 = 8

Only poorly formed/fused/cribriform glands, or

Predominantly well-formed glands and lesser component lacking glands, or

Predominantly lacking glands and lesser component of well-formed glands

Grade group 5: Gleason scores 9–10

Lack gland formation (or with necrosis) with or without poorly formed/fused/cribriform glands

From Epstein et al. (2016b).

Figure 2.

Adenocarcinoma of the prostate, Gleason grade 3 + 3 = score of 6 (grade group 1), with single, separate, well-formed glands in needle core biopsy from prostate.

Figure 3.

Adenocarcinoma of the prostate, Gleason grade 4 + 4 = score of 8 (grade group 4), with cribriform and fused glands in needle core biopsy from prostate.

Figure 4.

Adenocarcinoma of the prostate, Gleason grade 5 + 5 = score of 10 (grade group 5), with solid sheet-like growth in needle core biopsy from prostate.

Loss of basal cells is the second major defining attribute of invasive adenocarcinoma. Normal basal cells separate secretory luminal cells from the basement membrane. They frequently appear as rounded or oblong cells with small hyperchromatic nuclei and scant cytoplasm. Basal cells in prostatic glands may, however, be difficult to distinguish from periglandular fibroblasts and myofibroblasts, based on examination of H&E-stained sections. Furthermore, prostatic adenocarcinoma cells that are in thick histologic sections, are poorly fixed and preserved, and are distorted or crushed, may mimic basal cells. In diagnostically challenging situations, such as the diagnosis of small foci of adenocarcinoma (limited or minimal adenocarcinoma in needle biopsy tissue) or deceptively benign-appearing adenocarcinomas, immunohistochemistry to detect basal cell–specific proteins can be extremely useful (Hameed and Humphrey 2005; Epstein et al. 2014b). Basal cell loss is not absolutely specific for invasive adenocarcinoma, because some benign glands may lack basal cells (see section below on basal cell immunohistochemistry).

Nuclear atypia in the form of nuclear enlargement and nucleolar enlargement is the third of the major criteria. Nuclear atypia in malignant glands is most often manifested as nuclear enlargement and prominent nucleoli. A classical appearance of a prostatic carcinoma nucleus is a single large nucleus with chromatin clearing and a deeply staining macronucleolus. Although macronucleoli are characteristic of prostatic carcinoma, not all prostatic carcinomas harbor prominent nucleoli. How to define “prominence” is a matter of some debate and nucleoli with a diameter from 1 µm to 3 µm or greater have been deemed prominent in the past. In one study of limited carcinoma in needle biopsy, fully one quarter of the cases lacked prominent nucleoli (Epstein 1995). Additional carcinomas that may have small nuclei without prominent nucleoli include foamy gland carcinoma (Humphrey 2012) and carcinomas with androgen deprivation or radiation therapy effect (Srigley et al. 2012).

Minor, ancillary criteria can be helpful in diagnosing prostatic adenocarcinoma (Table 1). These minor characteristics are not specific for carcinoma but are useful in prompting in-depth study of the glands harboring these changes with a view to assessment of the above-discussed major diagnostic criteria. They include intraluminal mucin, pink amorphous intraluminal secretions, mitotic figures, cystalloids, and amphophilic cytoplasm.

Intraluminal wispy blue mucin identified on H&E-stained sections is more often found in carcinoma than in benign epithelial proliferations (Epstein 1995; Kaleem et al. 1998; Varma et al. 2002). The incidence in adenocarcinoma ranges from 18% to 34% in small foci in needle biopsies (Epstein 1995) to 72% in whole prostate glands at radical prostatectomy (Kaleem et al. 1998). This wispy blue mucin corresponds to acidic mucin, as substantiated by Alcian blue histochemical staining at pH 2.5. It is not specific for carcinoma by H&E or Alcian blue staining and can be recognized in mucinous metaplasia, atrophy, basal cell hyperplasia, sclerosing adenosis, atypical adenomatous hyperplasia (adenosis), and PIN.

Pink amorphous intraluminal secretions are also encountered more often in malignant than in benign glands (Epstein 1995; Kaleem et al. 1998; Thorson et al. 1998; Varma et al. 2002). These secretions are found in 53%–92% of adenocarcinomas in needle biopsy and in 84% of adenocarcinomas in the whole gland. They can be observed in a majority of high-grade PIN cases. They are discerned in only 2%–3% of benign glands, including 3% of benign atrophic glands. Scattered nuclear debris within the eosinophilic secretions should not be misinterpreted as necrotic tumor debris of comedocarcinoma.

Mitotic figures are an atypical finding in prostatic epithelium, with a trend toward an increased frequency in glands of carcinoma, being detectable in 0.001% of cells in benign glands versus 0.1% of cells in high-grade PIN versus 0.06%–0.15% of adenocarcinoma cells (Giannulis et al. 1993). Mitotic figures can be identified in 2%–11% of cases of minimal adenocarcinoma in needle biopsy and 38% of cases of adenocarcinoma in the whole gland (Epstein 1995; Kaleem et al. 1998; Thorson et al. 1998). Mitotic figures are not specific for adenocarcinoma and can occur in benign epithelium, especially adjacent to infarcts. The discovery of atypical mitotic figures is vanishingly rare in primary prostatic carcinoma, except in Gleason pattern 5 or neuroendocrine carcinoma. The diagnostic use of mitoses is diminished because of the lack of specificity, the more common occurrence in readily diagnosed higher-grade carcinoma, and the fact that most prostatic carcinomas in needle biopsy do not harbor mitotic figures.

Prostatic crystalloids are intraluminal, brightly eosinophilic, nonbirefringent structures that often assume geometric shapes such as needle-like, triangular, and hexagonal configurations (Jensen et al. 1980; Ro et al. 1986; Henneberry et al. 1997). They are found more often in adenocarcinoma than in benign glands. The incidence of crystalloids in adenocarcinoma is dependent on the extent of sampling and the histological grade of the adenocarcinoma, with the highest likelihood of detection in the lowest Gleason patterns 1 to 3 (Ro et al. 1986; Epstein 1995). Crystalloids are rarely found in high-grade Gleason pattern 4 or 5 adenocarcinomas. Crystalloids should be distinguished from intraluminal corpora amylacea, which are concentrically lamellated bodies that are, conversely, more often seen in benign (78% of cases) than in carcinomatous glands (11%–13% of cases in the whole gland and <1% in needle cores) (Humphrey and Vollmer 1990; Christian et al. 2005). Diagnostically, the prime advantage to the detection of crystalloids is to direct further scrutiny of the glands that contain the crystalloids.

The appearance of the cytoplasm in prostatic epithelial cells is of only slight assistance in the diagnosis of carcinoma. An amphophilic cytoplasm is more often present in carcinoma cells than in benign glandular cells, which tend to have a granular cleared cytoplasm. However, this is not specific and a cleared cytoplasm is common in low-grade Gleason pattern 1 and 2 prostatic adenocarcinomas. Furthermore, benign glands can have amphophilic cytoplasm in an appreciable number of cases. As for sensitivity, 36%–60% of prostatic carcinomas in needle biopsy and up to 100% of carcinomas in radical prostatectomy specimens have amphophilic cytoplasm (Epstein 1995; Kaleem et al. 1998; Thorson et al. 1998). In addition to being amphophilic or cleared, prostatic carcinoma cytoplasm can uncommonly reveal Paneth-cell-like neuroendocrine differentiation (Adlakha and Bostwick 1994; Epstein et al. 2014a; Park et al. 2014), xanthomatous cytoplasm as seen in foamy gland carcinoma (Nelson and Epstein 1996; Humphrey 2012), oncocytic cytoplasm (Pinto et al. 1994), and signet ring-like change (Humphrey 2012). None of these cytoplasmic changes is specific for malignancy.

The amount of cytoplasm can also be somewhat helpful in differential diagnosis of benignancy versus malignancy because prostatic carcinomas cells typically have a moderate amount of cytoplasm. In contrast, the benign entity most likely to be misdiagnosed as carcinoma—benign atrophy—has a scanty amount of cytoplasm. Amount of cytoplasm is not specific for a particular epithelial proliferation; however, as benign prostatic epithelial cells often have the same amount of cytoplasm as prostatic carcinoma cells and, moreover, the volume of prostatic carcinoma cytoplasm is low in atrophic pattern prostatic carcinoma.

Cellular characteristics that are not included in the diagnostic criteria checklist include cell size and shape and cytoplasmic borders. Carcinoma cellular borders can be distinct or not and the luminal border of gland-forming cells, although often flat, can be ruffled with bleb or snout formation.

Histopathological Features Considered Specific for Carcinoma

Histological features that have been considered specific for a diagnosis of prostatic carcinoma include extraprostatic spread of glands and, within the prostate, the findings of perineural invasion, collagenous micronodules, and glomeruloid intraglandular projections (Baisden et al. 1999).

The presence of prostatic glands outside the prostate is usually indicative of malignancy but it is important to know that ectopic benign prostatic glands have been found in many anatomic sites outside the prostate, including testis, epididymis, bladder, penile urethra, seminal vesicle, root of the penis, subvesical space, retrovesical space, pericolonic fat and submucosa, perirectal fat, urachal remnant, and spleen (Humphrey 2003b).

Perineural invasion is a hallmark of prostatic carcinoma. In whole prostate glands, perineural invasion can be found in the vast majority of cases with the range in the literature being 84%–94% (Byar and Mostofi 1972; Bostwick 1994). In needle core biopsy, about one-quarter of cases show perineural invasion, although, in a screening population with smaller lower-stage tumors, only 11% of cases had perineural invasion (Bismar et al. 2002). The diagnostic use of perineural invasion is diminished for a limited or minimal amount of carcinoma (<1 mm in greatest dimension) in needle biopsy tissue, because the incidence decreases to as low as 2% in this setting (Epstein 1995; Thorson et al. 1998). The mere presence of prostatic glands immediately adjacent to nerve is not, however, definitively diagnostic of malignancy. Benign prostatic glands can abut or wrap around nerves (Ali and Epstein 2005). One should rely on cytologic features of the epithelial cells around a nerve to distinguish benign from malignant perineural epithelium.

Lymphovascular space invasion by prostatic epithelial cells can be also viewed as specific for a diagnosis of malignancy. This finding is, however, rare in needle core tissue. Lymphovascular invasion can be found in 5%–53% of radical prostatectomy cases. Intraprostatic lymphovascular invasion is associated with higher grade, volume, and stage, and is related to increased risk of biochemical failure, distant metastases, and overall survival after radical prostatectomy (Magi-Galluzzi et al. 2011; Fajkovic et al. 2016). True lymphovascular invasion should be distinguished from artifactual separation of malignant glands from stroma, impingement of cancer on vascular spaces, and displacement of benign glands into lymphovascular spaces (Kryvenko and Epstein 2012); this may require immunohistochemistry for endothelial cells, using CD31 or podoplanin (D2-40) antibodies.

Collagenous micronodules, also known as mucinous fibroplasia, are microscopic aggregates of hyalinized stroma that represent an unusual response to invasive adenocarcinoma of the prostate (McNeal et al. 1991; Bostwick et al. 1995; Baisden et al. 1999). They are often, but not always, associated with abundant mucin production. The micronodules are mainly composed of collagen and are paucicellular, with a few elongated fibroblastic nuclei seen. The collagen can form true nodules, vaguely lobulated masses, and streaks and strands of the collagenous tissue within mucinous pools. The micronodules can be located in stroma adjacent to adenocarcinomatous glands and also within glandular lumina. For diagnostic purposes, collagenous micronodules seem to be highly specific for carcinoma, but their overall diagnostic use is somewhat limited, because they are found in only 1%–2% of needle biopsies with carcinoma (Bostwick et al. 1995; Thorson et al. 1998; Varma et al. 2002) and 13%–22% of carcinomas in whole glands in radical prostatectomy specimens (Bostwick et al. 1995; Kaleem et al. 1998). They are associated with Gleason pattern 3 or 4 adenocarcinoma (Kim et al. 2015).

Glomeruloid structures, or glomerulations, are renal glomerulus-like epithelial aggregates within acini of prostatic adenocarcinoma (Baisden et al. 1999). This pattern of intraluminal epithelial growth is thought to be specific for carcinoma (Pacelli et al. 1998). The diagnostic value of glomerulations is somewhat limited by their presence in just 3%–15% of needle biopsies with adenocarcinoma and 5% of radical prostatectomies with adenocarcinoma (Pacelli et al. 1998; Varma et al. 2002). Microscopically, glomeruloid bodies are characterized by rounded to ball-like tufts or buds into small to medium-size malignant glands. The glomeruloid bodies typically comprise only a minority of the carcinoma. They represent a high-grade Gleason pattern 4 (Epstein et al. 2016b).

Diagnostic Immunohistochemistry

Currently, the most valuable adjunctive technique for the diagnosis of adenocarcinoma of the prostate is immunohistochemistry with antibodies directed against basal cells (34βE12 and p63) and α-methylacyl-CoA racemase, or AMACR (also known as P504S and racemase) (Hameed and Humphrey 2005; Brimo and Epstein 2012; Epstein et al. 2014b). The best characterized basal cell immunostains use monoclonal antibody 34βE12, which binds to high-molecular-weight cytokeratins expressed in the cytoplasm of basal cells and not luminal cells, and an antibody directed against p63, which is localized to basal cell nuclei. Cytokeratin 5/6 may also be used as a basal cell marker. A cocktail of p63 and 34βE12 antibodies appears to increase sensitivity of basal cell detection (Zhou et al. 2003a). It is important to be aware that some benign glands can have a discontinuous or absent basal cell layer. Indeed, one of the benign entities most often misdiagnosed as carcinoma—atrophy—can reveal scattered basal cell–negative glands in up to a quarter of cases (Hameed and Humphrey 2005). Also, on average, 50% (range: 10% to 90%) of glands in atypical adenomatous hyperplasia (adenosis) do not stain for basal cells. Thus, diffuse absence of basal cells in the glands of concern by immunohistochemistry is most supportive, in the correct histological context, of a diagnosis of adenocarcinoma. Finally, there are exceedingly rare, molecularly distinctive, invasive adenocarcinomas that are p63 immunoreactive but negative for high molecular weight cytokeratins (Osunkoya et al. 2008a; Giannico et al. 2013; Tan et al. 2015).

AMACR immunostaining, for overexpression of this enzyme, has also assumed a prominent role as a confirmatory marker for prostatic carcinoma (Jiang et al. 2001; Rubin et al. 2002; Epstein et al. 2014b). An advantage of the AMACR immunostain is that it is a positive signal for neoplastic prostatic epithelial cells. This marker was discovered to be overexpressed in prostatic adenocarcinoma by cDNA library subtraction using high-throughput DNA microanalysis (Xu et al. 2000). In immunohistochemistry, this enzyme is selectively expressed in 80%–100% of prostatic carcinomas but it is not specific for carcinoma. Up to 20% of benign glands can also stain, but the staining is weaker and more focal than in carcinoma. Circumferential, granular cytoplasmic staining in prostatic carcinoma glands is characteristic. It is important to run an immunostain for basal cell marker(s) at the same time as the AMACR immunostain, because high-grade PIN and atypical adenomatous hyperplasia (adenosis) can also be AMACR-positive (but differ from carcinoma in having a least-fragmented basal cell layer). AMACR positivity can also be seen in atrophy (especially partial atrophy) and nephrogenic adenoma. Basal cell marker and AMACR immunostains can be run separately on different sections or slides on a single section or slide as a cocktail of p63/AMACR or as the so-called “triple stain” using three antibodies—34βE12/p63/AMACR—and two chromogens, if only a few glands are available for evaluation (Jiang et al. 2005).

ERG expression, as detected by immunohistochemistry, is fairly specific for adenocarcinoma of the prostate (and vascular tumors) (Park et al. 2010; Miettinen et al. 2011) but its usage as a diagnostic marker is limited because of the relative lack of sensitivity with only ∼50% of prostatic adenocarcinomas being positive. It does not provide added value beyond basal cell markers and AMACR in diagnosis of minimal adenocarcinoma (Andrews and Humphrey 2014).

The differential diagnosis of high-grade adenocarcinoma of the prostate versus high-grade urothelial carcinoma can be aided by use of a targeted panel of antibodies in immunohistochemistry, with use of PSA, prostatic acid phosphatase (PAP), and NKX3.1 or prostein antibodies for prostatic adenocarcinoma and with usage of GATA3, p63, and 34βE12 antibodies for urothelial carcinoma (Epstein et al. 2014b). PSA and GATA3 have been recommended as first-line markers (Epstein et al. 2014b).

The distinction of high-grade prostatic adenocarcinoma versus urinary bladder adenocarcinoma can be addressed with the prostatic markers PSA, PAP, and prostein. Villin, thrombomodulin, CDX2, and carcinoembryonic antigen (CEA) immunostains can be applied as urinary bladder adenocarcinoma markers (Epstein et al. 2014b).

Colonic adenocarcinoma can be distinguished from high-grade prostatic adenocarcinoma by use of PSA, PAP, prostein, and NKX3.1 immunostains (for prostatic differentiation) compared with expression of villin and CDX2 in colonic adenocarcinoma (Epstein et al. 2014b).

The diagnostic separation of small seminal vesicle glands and prostatic adenocarcinoma can usually be achieved by examination of H&E-stained slides. When necessary, this can also be accomplished by PSA, PAP, and 34βE12/p63/AMACR immunohistochemistry in which prostatic carcinoma glands will be positive for PSA, PAP, and AMACR and negative for basal markers with seminal vesicle glands showing the opposite immunophenotype (Harvey et al. 2010).

The tissue diagnosis of metastatic adenocarcinoma can be confirmed by positive PSA, PAP, prostein, and NKX3.1 immunostains. Each of these four prostatic markers displays >94% sensitivity (Epstein et al. 2014b). PSA and PAP expression can be down-regulated with androgen deprivation therapy and NKX3.1 and prostein may be useful in this scenario. As far as specificity, PSA expression can be observed in salivary gland neoplasms and PAP immunoreactivity may be detected in salivary gland and neuroendocrine neoplasms. NKX3.1 and prostein are highly specific for metastatic adenocarcinoma of the prostate (Sheridan et al. 2007; Gurel et al. 2010). Markers that are not recommended for diagnosis of metastatic adenocarcinoma of the prostate in metastatic sites include prostate-specific membrane antigen, ERG, androgen receptor, and AMACR (Epstein et al. 2014b).

Variants of Acinar Adenocarcinoma

Variants of usual acinar adenocarcinoma include, according to the 2016 World Health Organization (WHO) scheme, the following: atrophic, pseudohyperplastic, microcystic, foamy, mucinous (colloid), signet ring-like cell, pleomorphic giant cell, and sarcomatoid (Table 3) (Humphrey et al. 2016a). The first four variants (atrophic, pseudohyperplastic, microcystic, and foamy) are of significance because they are deceptively benign-appearing and may be difficult to diagnose. The last three variants (signet ring-like cell, pleomorphic giant cell, and sarcomatoid) harbor prognostic significance, with a worse prognosis compared with usual acinar adenocarcinoma.

Table 3.

2016 World Health Organization (WHO) classification of carcinomas and neuroendocrine tumors of the prostate

Glandular neoplasms

Acinar adenocarcinoma

Atrophic

Pseudohyperplastic

Microcystic

Foamy gland

Mucinous (colloid)

Signet ring-like cell

Pleomorphic giant cell

Sarcomatoid

Intraductal carcinoma

Ductal adenocarcinoma

Urothelial carcinoma

Squamous neoplasms

Adenosquamous carcinoma

Squamous cell carcinoma

Basal cell carcinoma

Neuroendocrine tumors

Adenocarcinoma with neuroendocrine differentiation

Well-differentiated neuroendocrine tumor

Small-cell neuroendocrine carcinoma

Large-cell neuroendocrine carcinoma

Atrophic pattern adenocarcinoma may be seen in sporadic and postradiation or postandrogen deprivation therapy settings and is characterized by cytoplasmic volume loss, just like benign atrophy (Cinha and Epstein 1997; Egan et al. 1997; Reuter 1997; Kaleem et al. 1998). An infiltrative architecture, macronucleoli, nucleomegaly, and admixed nonatrophic usual acinar adenocarcinoma with a moderate amount of cytoplasm are useful findings in diagnostic recognition. Challenges in diagnosis include identification in some glands of nuclear atypia due to nuclear compression and diminished expression of AMACR, with 70% of cases being positive (Farinola and Epstein 2004). Complete loss of basal cells in all glands is seen in immunohistochemistry. Most atrophic pattern adenocarcinomas are Gleason pattern 3. The presence of atrophic features in adenocarcinoma is not likely of prognostic significance because adenocarcinomas with and without atrophic change do not differ in Gleason grade or pathologic stage (Kaleem et al. 1998).

Pseudohyperplastic adenocarcinoma can simulate the appearance of usual epithelial hyperplasia with papillary infoldings, luminal undulations and branching, and cystic dilatation (Humphrey et al. 1998; Wolf and Epstein 2000; Arista-Nasr et al. 2015). AMACR expression is detectable in 70% to 83% of cases (Zhou et al. 2003b) and no basal cells are found in the glands of concern by immunostaining. Pseudohyperplastic adenocarcinomas are Gleason pattern 3. The presence of pseudohyperplastic change does not impact pathological stage (Humphrey et al. 1998). Of note, HOXB13 G84E-related familial prostate cancers commonly show pseudohyperplastic features (Smith et al. 2014).

Microcystic adenocarcinoma shows gland dilatation with intermediate-sized glands that are, on average, 10 times the size of usual small acinar adenocarcinoma glands (Yaskiv et al. 2010). The incidence in radical prostatectomy cases is 11%. Microscopically, the expansion of the luminal spaces generates a rounded profile, and the luminal cell-lining layer is flat with or without atrophic changes. Intraluminal crystalloids and blue mucin are uniformly present. In immunohistochemistry, the dilated glands lack basal cells and almost all cases (96%) express AMACR (Yaskiv et al. 2010). The Gleason pattern assignment is 3.

Foamy gland adenocarcinoma is characterized by abundant foamy or xanthomatous-type cytoplasm with pyknotic nuclei being frequent (Nelson and Epstein 1996; Tran et al. 2001; Koca et al. 2014). Foamy glands are observed admixed with usual acinar adenocarcinoma in 17% of needle biopsy cases (Warrick and Humphrey 2013) and 13%–23% of radical prostatectomy cases (Hudson et al. 2012) and is only rarely found in pure form. The lack of nuclear atypia can create difficulty in establishment of a malignant diagnosis, especially in needle core tissue. Most foamy gland carcinomas are Gleason score 6 or 7, although Gleason score 8 to 10 carcinomas also exist. Foamy gland carcinoma has a similar prognosis as nonfoamy gland adenocarcinoma (Hudson et al. 2012).

Mucinous (colloid) adenocarcinoma is defined as prostatic adenocarcinoma with at least 25% of the tumor comprised of lakes of extracellular mucin (Ro et al. 1990; Saito and Iwaki 1999). It constitutes ∼0.3% of all prostatic adenocarcinomas. Microscopically mucinous adenocarcinoma shows tumor cells “floating” or embedded within prominent pools of mucin. Most cases are Gleason score 7 with a minority being Gleason score 6 or 8 (Osunkoya et al. 2008b). In the past, mucinous adenocarcinoma was thought to confer a worse prognosis than usual acinar adenocarcinoma but more recent reports suggest that mucinous adenocarcinoma is not more aggressive (Lane et al. 2006; Osunkoya et al. 2008b; Marcus et al. 2012).

Signet ring-like cell carcinoma is rare and is microscopically characterized by carcinoma cells with nuclear displacement and indentation by clear cytoplasmic vacuoles (Torbenson et al. 1998; Humphrey 2012). In contrast to signet-ring carcinomas at other sites, intracellular mucin may not be detectable by histochemical mucin stains of prostatic signet ring–like cell carcinoma. These signet ring–like cells can grow as sheets, cords, small clusters, and as single cells. Immunostains for PSA and PAP are positive in most cases. The assigned Gleason pattern is 5. The outcome is poor with a mean survival of 28 months (Warner et al. 2010).

Pleomorphic giant cell carcinoma is extremely rare. Histological sections show giant pleomorphic nuclei with admixed high-grade usual acinar adenocarcinoma of Gleason score 9 (Lopez-Beltran et al. 2005; Parwani et al. 2006). These tumors are very aggressive.

Sarcomatoid carcinoma (also called carcinosarcoma) is a rare biphasic malignancy in the prostate with malignant epithelial and mesenchymal elements (Shannon et al. 1992; Dundore et al. 1995; Hansel and Epstein 2006). In about one-half of patients, the initial diagnosis was usual acinar adenocarcinoma, followed by hormonal and/or radiation therapy, with a subsequent diagnosis, after a mean of 7 years, of sarcomatoid carcinoma. Sarcomatoid carcinoma may be either homologous, in which the mesenchymal-like areas have the appearance of an undifferentiated sarcoma, or heterologous, in which the sarcoma shows differentiation along the lines of specific mesenchymal cells such as bone and cartilage. The carcinomatous portion is almost always glandular and acinar, whereas the sarcomatoid component often shows areas of undifferentiated spindled and pleomorphic cells arranged in sheets, pinwheels, or fascicles resembling pleomorphic undifferentiated sarcoma. Osteosarcoma and chondrosarcoma are by far the most frequently identified heterologous elements. The prognosis for these patients is poor, although recent data suggest that localized disease can be effectively treated with surgery and/or radiation therapy (Markowski et al. 2015).

VARIANTS/TYPES OF NONACINAR CARCINOMA

These prostatic carcinoma variants/types include, according to the 2016 WHO classification (Humphrey et al. 2016a), ductal adenocarcinoma, urothelial carcinoma, squamous neoplasms, basal cell carcinoma, and neuroendocrine tumors (Table 3).

Ductal adenocarcinoma is a subtype of adenocarcinoma of the prostate with typically large glands lined by tall pseudostratified columar cells (Fig. 5) (Egevad et al. 2016b; Seipel et al. 2016). Pure ductal and mixed ductal–acinar carcinomas comprise ∼1% and 5% of all prostatic carcinomas, respectively (Dube et al. 1973). Obstruction and hematuria are common clinical manifestations, unlike usual acinar adenocarcinoma. Macroscopically, at urethroscopy, there may be an exophytic, villous/polypoid growth protruding into the urethra at or near the verumontanum. Microscopically, papillary and cribriform architectures are often found, although solid and single separate PIN-like patterns are less common (Hameed and Humphrey 2006; Tavora and Epstein 2008). A prominent in situ neoplastic component is often seen. The immunophenotype and gene expression profile are similar to acinar adenocarcinoma (Sanati et al. 2009; Seipel et al. 2016). The proliferation index is higher in ductal adenocarcinoma than acinar adenocarcinoma but overlap precludes its use. No marker currently will distinguish acinar and ductal adenocarcinomas of the prostate. The spread of ductal adenocarcinoma is the same as for acinar adenocarcinoma, although ductal adenocarcinoma has a propensity to spread to testis, penis, and lung. In most studies, the outcome for men with pure or predominantly ductal adenocarcinoma is worse than for average acinar adenocarcinoma, likely the result of higher stage and grade. The Gleason grade of ductal adenocarcinoma is usually 4, but uncommonly patterns 3 (PIN-like pattern) or 5 (solid and comedo patterns) can be seen.

Figure 5.

Ductal type of adenocarcinoma of the prostate with papillary growth. This is a high-grade (Gleason score 8) prostate cancer.

Urothelial carcinoma in the prostate usually represents secondary involvement but can arise as a primary in the large, central, primary ducts of the prostate that are normally lined by urothelium (Greene et al. 1973; Cheville et al. 1998; Shen et al. 2006; Grignon 2016). Microscopically, prostatic urothelial carcinoma has a striking propensity for growth within prostatic ducts and acini with solid cylinders, with or without comedo necrosis, being common. Pagetoid and undermining spread of the neoplastic urothelial cells within benign prostatic epithelium may be visualized. Stromal invasion is typified by irregular nests and cords of pleomorphic tumor cells, which often have a squamoid, eosinophilic cytoplasm. As noted above, urothelial markers that can be useful in the distinction versus prostatic adenocarcinoma include thrombomodulin, GATA3, p63, and high molecular weight cytokeratins. Urothelial carcinomas are negative for PSA, PAP, NKX3.1, and prostein in immunohistochemistry. Outcome is dependent on pathologic stage (Shen et al. 2006; Grignon 2016).

Squamous and adenosquamous carcinomas primary in the prostate are exceedingly rare (Parwani et al. 2004; Marcus et al. 2012). Many are diagnosed after radiation or hormonal therapy. One should exclude secondary involvement of the prostate by bladder squamous carcinoma or urothelial carcinoma with squamous differentiation. Survival is, on average, 1 year.

Basal cell carcinoma is a malignant neoplasm of prostatic basal cells. They are extraordinarily rare tumors that can resemble, to a certain degree, adenoid cystic carcinomas of the salivary glands (Yang et al. 1998; Iczkowski et al. 2003; Ali and Epstein 2007; Simper et al. 2015). Microscopically, growth arrangements include adenoid cystic/cribriform patterns and small solid nests with peripheral palisading. Other patterns include a basal cell hyperplasia-like arrangement, small tubules with a hyaline rim, and large solid nests with or without necrosis. Bcl-2 expression and a higher Ki-67 proliferation index favor basal cell carcinoma more than basal cell hyperplasia (Yang et al. 1998). A subset of cases with the adenoid cystic-like pattern show MYB rearrangement (Bishop et al. 2015). Extraprostatic extension has been reported in a minority of cases and metastasis has been detected in ∼15% of cases. The solid nested pattern is associated with more aggressive behavior.

Neuroendocrine carcinomas in the prostate include adenocarcinoma with neuroendocrine differentiation, small-cell neuroendocrine carcinoma, and large-cell neuroendocrine carcinoma (Epstein et al. 2014a, 2016a). Usual acinar adenocarcinoma expresses neuroendocrine markers such as chromogranin, synaptophysin, and CD56 in 10% to 100% of cases. Immunostains for these markers are not recommended for routine use in diagnosis of typical adenocarcinoma. The significance of the expression of these markers is uncertain (Epstein et al. 2016a). Acinar adenocarcinoma may also show Paneth-like cells with neuroendocrine cytoplasmic granules. Small-cell carcinoma primary in the prostate is a rare, extremely aggressive malignancy that often presents with disseminated disease (Têtu et al. 1987; Wang and Epstein 2008; Epstein et al. 2014a, 2016a; Nadal et al. 2014). In about one-half of cases, the carcinoma is pure small-cell carcinoma and, in the other half, there is an admixture with prostatic acinar adenocarcinoma. An intriguing and important aspect of prostatic small-cell carcinoma diagnosis is that in one-third of patients there is an initial diagnosis of adenocarcinoma followed by therapy, usually hormonal, in turn followed (at a median of 18–49 months, range: 7 months to 8 years) by a subsequent diagnosis of small-cell carcinoma. Microscopically, prostatic small-cell carcinoma grows as sheets, with ribbons, nests, and occasional rosette-like structures noted. The admixed adenocarcinoma is variable in grade and extent. Proof of prostatic origin of metastatic small-cell carcinoma can be accomplished in 46% to 86% of cases by detection of TMPRSS2-ERG gene fusion by FISH (Scheble et al. 2010; Guo et al. 2011; Schelling et al. 2013). Neuroendocrine markers (synaptophysin, chromogranin, and CD56) are detected in ∼90% of cases. The prostatic markers PSA, PAP, PSMA, and prostein, in contrast, are identified by immunostaining in only a small minority of cases of prostatic small-cell carcinoma (Yao et al. 2006; Epstein et al. 2014b). The median average survival for patients with small-cell carcinoma of the prostate is 1 to 2 years. Large-cell neuroendocrine carcinoma of the prostate is extremely rare (Evans et al. 2006). Sections show large nests with peripheral palisading and sometimes geographic necrosis. The mean survival is 7 months.

TREATMENT EFFECTS

Grossly, the prostates of men treated with androgen deprivation therapy are small and atrophic. Microscopically, androgen deprivation therapy effects on prostatic adenocarcinoma include a decrease in number of malignant glands with loss of luminal glandular spaces (luminal collapse), increased stroma, nuclear pyknosis, and cytoplasmic vacuolization (Armas et al. 1994; Srigley et al. 2012). These treated carcinoma cells can resemble lymphocytes or histiocytes. Histologic grading of prostatic carcinoma after hormonal therapy is not recommended because the architectural changes of luminal collapse simulate high-grade carcinoma (Humphrey et al. 2016a). Immunostains for PSA, low molecular weight cytokeratins, basal cells, and AMACR may be of use in identification of histologically inconspicuous carcinoma cells with treatment effect. AMACR is, however, down-regulated by androgen deprivation therapy in some cases, with 45% to 71% of cases being positive (Suzue et al. 2005; Sung et al. 2007).

Hormonal treatment with 5α reductase inhibitors such as finasteride and dustasteride has a minimal-to-absent effect on prostatic carcinoma histomorphology and so Gleason grades may still be assigned after this type of treatment (Srigley et al. 2012). The finasteride-induced selective suppression of low-grade prostate cancer in the Prostate Cancer Prevention Trial may be due, in part, to the differential effect of finasteride on apoptosis and androgen receptor expression in low-grade (Gleason pattern 3) compared with higher-grade (Gleason pattern 4) tumors (Kim et al. 2016).

Radiation therapy–induced histological changes in carcinoma are a decrease in number of tumor glands, generation of poorly formed glands and single cells, cytoplasmic vacuolization, nuclear pyknosis, and stromal fibrosis (Cheng et al. 1999). Gleason grading should be attempted when there is no obvious radiotherapy effect but should not be performed when there is marked radiotherapy effect. Such severe radiation therapy histomorphologic effect on carcinoma cells should be reported because these patients have the same prognosis as patients without carcinoma (Crook et al. 2009). Immunohistochemistry for basal cells and AMACR can be of value in the differential diagnostic consideration of benign prostatic epithelial cells with radiotherapy-induced atypia, which can be striking, and in detection of carcinoma cells with marked radiotherapy effect.

Neoadjuvant chemotherapy with docetaxel, docetaxel with mitoxantrone, or abiraterone and enzalutamide induces inconspicuous collapsed glands, small inconspicous individual tumor cells, cytoplasmic vacuolization, and intraductal and cribriform growth patterns (Magi-Galluzzi et al. 2007; O’Brien et al. 2010; Murphy et al. 2016). A three-group system has been proposed to classify these histological alterations (Efstathoiu et al. 2010). Intraductal carcinoma and a cribriform growth pattern in radical prostatectomy tissue sections (group C tumors) are associated with adverse clinical outcome after neoadjuvant chemotherapy (Efstathoiu et al. 2010; O’Brien et al. 2010).

Cryosurgery can cause necrosis and nonspecific halinization, fibrosis, calcification, and hemosiderin deposition (Borkowski et al. 1996). High-intensity ultrasound treatment may produce fibrosis, inflammation, and necrosis (Biermann et al. 2010; Ryan et al. 2012). No significant therapy-specific changes in the residual carcinoma are identified. Gleason grading may be performed for residual carcinoma detected after these treatments.

GRADING OF PROSTATE CANCER

The Gleason grading system is currently the most widely used histologic grading scheme for prostatic adenocarcinoma in the United States and worldwide (Gleason 1966, 1992; Gleason et al. 1974, 1977; Humphrey et al. 2016a,b). The Gleason grading method is based entirely on architectural arrangements of prostatic carcinoma (Fig. 1). The grading diagram depicting Gleason patterns has undergone modifications from the original diagram. This most current diagram is based on modifications according to a 2014 consensus meeting of the ISUP published in 2016 (Epstein et al. 2016b) and was endorsed by the World Health Organization (Fig.1) (Humphrey et al. 2016a,b).

In the Gleason system, the histologic patterns are categorized into five basic grade patterns at relatively low magnification (using a 4× or 10× objective) by the extent of glandular differentiation and the pattern of growth of the tumor in the prostatic stroma. These five basic grade patterns are used to generate a histologic score, which can range from 2 to 10, by adding the primary grade patterns and the secondary grade pattern. The primary pattern is the one that is predominant in area on simple visual inspection. The secondary pattern is the second most common pattern. If only one grade is in the tissue sample, then that grade is multiplied by two to give the score. If, however, high-grade (pattern 4 or 5) carcinoma is present, and a low-grade pattern occupies <5% of the tumor, then the low-grade component should not be included in the score. For example, if a tumor is 96% pattern 4 and 4% pattern 3, then the score should be 4 + 4 = 8. In needle biopsy samples, a high-grade pattern of any amount should always be incorporated into the score. So, in needle biopsy of 96% pattern 3 and 4% pattern 4, the score is 3 + 4 = 7. A Gleason grade (primary + secondary = score) should be provided for all primary adenocarcinomas diagnosed in prostatic tissue, even minimal adenocarcinoma in needle biopsy tissue. The only setting in which a prostatic adenocarcinoma should not be assigned a Gleason grade is when there is a hormonal or a radiation therapy effect. Note that Gleason grading is applicable only to adenocarcinomas within the prostate gland.

The Gleason grading system allows for two separate grade patterns in an individual tissue sample, but the histomorphological appearance of prostatic carcinoma is more heterogeneous than this. Indeed, in one study, an average of 2.7 Gleason grade patterns (range 1–5) was found in carcinomas in whole prostate glands (Aihara et al. 1994). The number of grades assigned depends on tumor sample size and size of the tumor in the whole gland. In needle biopsy tissue, 4% of cases have more than two grades, whereas tumors >1–2 cm³ in size in radical prostatectomy tissue sections tend to have more than two grades (Aihara et al. 1994; Ruijter et al. 1996).

Tertiary grades are handled differently depending on the tissue sample. For needle biopsies with patterns 3, 4, and 5, the score is assigned using the predominant pattern and the highest grade (Epstein et al. 2005; Epstein 2010). So, for patterns 4 + 3 + 5, in order of pattern amount, the Gleason score is 4 + 5 = 9. For radical prostatectomy specimens, the presence of a third pattern that is <5% of the tumor and that is higher in grade than the primary and secondary grades should be noted in the report rather than incorporated into the score. So 3 + 4 + 5, in order of pattern amount in radical prostatectomy should be diagnosed as 3 + 4 = score of 7 with a tertiary pattern 5. This is so because the outcome for these patients is intermediate between score of 7 and score of 8. It is clear that a high-grade Gleason pattern 4 or 5 that is a tertiary component occupying <5% of the tumor in the whole gland influences pathological stage and progression rates (Pan et al. 2000; Mosse et al. 2004; Hattab et al. 2006).

Comparison of needle biopsy Gleason grade with grade in the matched whole glands reveals reasonable agreement, but undergrading is common. Intraobserver and interobserver variability in Gleason grade assignment does exist (Allsbrook et al. 2001a), but a substantial degree of interobserver agreement can be achieved (Allsbrook et al. 2001b). For needle biopsy cases, there are several sources of grading error, including difficulty in appreciation of an infiltrative growth pattern, tissue sampling error (related to the small amount of tissue supplied by needle biopsy and grade heterogeneity), tissue distortion, pathologist experience, and observer variability.

Percentage of prostate cancer that is high-grade Gleason pattern 4 or 5 is an important prognostic indicator (McNeal et al. 1990; Stamey et al. 1999; Egevad et al. 2002; Cheng et al. 2007). The 2016 WHO Classification of Tumors of the Urinary System and Male Genital Organs recommends reporting percentage pattern 4 for Gleason score 7 adenocarcinomas (Humphrey et al. 2016a). Risk stratification based on percentage pattern 4 has been shown for Gleason score 7 adenocarcinoma in needle biopsy (Cole et al. 2016) and radical prostatectomy (Choy et al. 2016; Sauter et al. 2016) tissues. Knowledge of the percentage pattern 4 in Gleason score 7 adenocarcinoma in needle biopsy tissue may be useful in selection of patients for active surveillance (Morash et al. 2015; Chen et al. 2016). In particular, select patients with clinically localized Gleason grade 3 + 4 = score of 7 adenocarcinoma with ≤10% pattern 4 may be candidates for active surveillance (Morash et al. 2015). Assessment of percentage Gleason pattern 4 is relatively reproducible (Sadimin et al. 2016), although it may be a challenge to visually estimate parentage pattern 4 when patterns 3 and 4 are intermingled, which is a common occurrence (Egevad et al. 2016a).

Grade groups should be reported along with the 2014 WHO/ISUP-modified Gleason score (Epstein et al. 2016b; Humphrey et al. 2016a). These grade groups are defined in Table 2 (Pierorazio et al. 2013). A rationale for their use is that Gleason score 6 of 10, which suggests intermediate grade, is now more appropriately categorized as the lowest grade group 1 of 5, thereby informing physicians and patients of the relatively indolent nature of this large group of prostate cancers (Epstein et al. 2016b). Additionally, the grading is simplified into five categories rather than 12, and the grade groups appear to provide better stratification of risk compared with standard Gleason scoring, using Gleason score categories of 6 versus 7 versus 8–10 (Spratt et al. 2016a,b). Prognostic use of the grade groups, which have been termed ISUP grade in some publications, has been validated in a number of studies (Delahunt et al. 2015; Samaratunga et al. 2015; Berney et al. 2016; Epstein et al. 2016d; Spratt et al. 2016a,b). Of interest, there are genomic correlates to and molecular support for the grade group system (Rubin et al. 2016).

HISTOPATHOLOGIC APPEARANCE OF PROSTATIC ADENOCARCINOMA THAT IS LOCALLY ADVANCED OR METASTATIC

The histopathologic appearance of prostatic adenocarcinoma that has spread outside the prostate gland often resembles one of the patterns in the primary tumor and is usually high-grade Gleason pattern 4 or 5. Gleason grading is not, however, applied to prostatic adenocarcinoma that is outside the prostate. Diagnostic difficulty can arise in patients with metastatic prostate cancer treated with androgen deprivation agents when the metastatic adenocarcinoma cells show hormonal treatment effect such that the malignant cells appear benign and morphologically simulate histiocytes.

When a man presents with metastatic adenocarcinoma of unknown primary origin, the histomorphological appearance of a carcinoma with uniform nuclei and prominent nucleoli may suggest a prostatic primary but immunohistochemical stains for PSA, PAP, and NKX3.1 should always be performed to support a diagnosis of metastastic adenocarcinoma of the prostate.

CONCLUDING REMARKS

Histopathologic assessment of H&E-stained tissue sections with use of immunohistochemistry in select cases is currently the standard approach in diagnosis and characterization of prostate cancer. In the future, additional molecular markers beyond the proteins currently targeted in immunohistochemistry and including RNA expression profiles and DNA abnormalities may be of use in tissue diagnosis of prostate cancer and in risk stratification of patients with prostate cancer.